Chemical shifts of the olefinic carbons of o-substituted 1-pentenes and 1-butenes and the aromatic carbons of phenylethanes and phenylpropanes have been measured. Correlations with various substituent constants, electronegativities, NMR data from related chemical systems and calculated electrical fields were evaluated. Although the pattern of results is quite consistent with expectations for a field-induced shift of =-electron density, simple correlations of high statistical precision are not found. The general problem of correlating substituent effects with reactivity and NMR shifts is discussed.In the course of other studies, we observed that the chemical shifts of the olefinic carbons of 5-bromo-lpentene differ from those of its Grignard reagent by several parts per million. Since the position of substitution is 8 and E to the affected carbons, the substituent effects appeared to be rather larger than expected. We therefore studied a number of related compounds. We have concluded that the magnitude of the substituent effects is indeed 'normal', resulting from .rr-electron polarization by the polar substituent groups, and does not represent a specific interaction unique to these particular substituents. A number of other recent studies of NMR substituent effects also support the concept of bond electron polarization by polar substituents.'-'6 In this paper we report the chemical shifts of the olefinic carbons of a number of w-substituted 1-butenes and 1-pentenes (1, n = 2, 3), the aromatic carbons of the phenyl analogs (2, n = 2, 3) and olefinic carbons of some related cyclic compounds, and our attempts to correlate these shifts with measures of polar substituent effects.
RESULTSIn Tables 1 and 2, chemical shifts are reported for series 1 and 2, respectively. Spectra were run mostly at concentrations of 10-15% by volume in diethyl ether. Shifts for some compounds were concentrationdependent at higher concentrations, but little change appeared to result from further dilution. Omitted from correlations because of poor fit to plot of 6(C-1) vs 6(C-2).